Workpackage 2. Indicators for the Knowledge-Based Economy: Summary Report. Deliverable 2.5

Transcription

1 Workpackage 2 Indicators for the Knowledge-Based Economy: Summary Report Deliverable

2 List of contributors: 1 Anthony Arundel, Wendy Hansen, Minna Kanerva, MERIT Main responsibility: Anthony Arundel, Wendy Hansen, Minna Kanerva, MERIT C1S8-CT KEI The project is supported by European Commission by funding from the Sixth Framework Programme for Research We would like to acknowledge comments and input received from Asterios Chatziparadeisis for the revision of this document. ii KEI-WP2-D2.5

3 Table of Contents 1. INTRODUCTION Overview of indicators INDICATORS Group A: Characteristics and drivers A1 - Production and diffusion of ICT A2 - Human resources, skills and creativity A3 - Knowledge production and diffusion A4 - Innovation, entrepreneurship and creative destruction Group B: Outputs - Economic performance, society and the KBE B1 - Economic outputs B2 - Social performance Group C: Globalisation C1a - Trade C1b - Knowledge production and diffusion C1c - Economic structure C1d Human resources CONCLUSIONS AND FURTHER THOUGHTS Overall conclusions Observations on indicator availability for the KBE ANNEX I iii

4

5 1. INTRODUCTION The Knowledge Economy Indicators (KEI) project evaluated hundreds of indicators for their usefulness in evaluating and tracking the development of a Knowledge Based Economy (KBE) in Europe and among other major countries outside Europe. Virtually hundreds of indicators are of potential use, but many indicators suffer from problems of availability and consistency across Europe and for countries of interest outside of Europe. This report summarizes research to evaluate current and potential indicators for measuring the drivers, characteristics and key outputs of a KBE. The project assembled a dataset of relevant indicators using currently available data. 2 The report summarizes the state-of-the-art for indicators, identifies the types of indicators that are available for defining the characteristics of a KBE, and identifies the drivers of growth, performance and output of a KBE. The rest of this introduction provides an overview of the KEI classification system for KBE indicators and discusses the policy and user needs for KBE indicators, drawing from a round of interviews with policy analysts. Section 2 provides the core of this report and gives details on the KBE indicators. Conclusions are provided in Section 3. Annex I presents a table of the 64 selected, currently available key KBE indicators. 1.1 Overview of indicators The purpose of the KEI project is to identify key indicators for knowledge economies and methodologies for constructing composite indicators to measure and compare national KBE performance. The latter include performance for innovation (e.g. capabilities and output) and globalisation (e.g. impact of globalisation on work and life). Of note, the inputs for a KBE are not limited to R&D, but include many other types of activities. Similarly, the outputs should not be measured only in economic terms, but should include quality of life issues. Both innovation and the drivers of a knowledge economy require a broad set of skills and capabilities. One set of indicators of particular policy relevance for gauging the knowledge capabilities of a nation (and one used for composite indicators on human development) has been the supply of scientists and engineers (S&E) and student performance in mathematics and science. In some cases this focus has led to the neglect of other valuable skills for innovation, including reading literacy and communication skills. This is a mistake the transition of work and life brought about by the KBE requires networking, problem solving and communication skills, from writing to marketing. Consequently, the indicators for a KBE need to consider reading and writing as important as mathematics and science. 2 Key missing indicators are discussed in WP4, which also assesses future data needs for a KBE

6 What are some of the relevant indicators needed for constructing indicators for the KBE? Table 1 in Annex I provides a detailed overview of 64 relevant indicators that were chosen from the pool of currently available indicators. Due to data limitations, they represent a conservative viewpoint on how best to measure the KBE. The KBE indicators are classified into four main drivers (Group A), two main outputs (Group B), and a number of horizontal indicators of globalisation (Group C). The Group A indicators on the drivers and characteristics of a KBE are further divided into four main sub-groups: Production and diffusion of ICTs. ICT is the main technology underlying the KBE and increases in productivity. 3 Human resources, skills and creativity. These indicators represent the primary set of indicators needed to develop composite indicators on the human potential of a nation. These indicators reveal the creative and absorptive capacity of a work force. Knowledge production and diffusion. This group includes many of the traditional indicators of R&D and knowledge production. They provide us with sound trend data and with indicators to develop composite measures of globalisation and competitiveness. Innovation, entrepreneurship and creative destruction. These indicators tell us about the churn and change brought about by ICTs and globalising knowledge economies. They cover firm behaviour and aspects of innovation including demand for innovative products, financing and market innovation. The share of the population in specific age cohorts can provide insights into the demand for innovative products. 4 Group B includes output indicators, from economic outputs to measures of social performance: Economic output. This group of indicators includes a number of horizontal basic indicators, such as GDP, value added and employment that can be used to develop indicators of economic outputs including income measures, productivity measures and employment trends. Social performance. These indicators cover a broader landscape of policy issues in the KBE. They can be used to develop composite indicators on the environment and sustainable growth, economic welfare and quality of life. Lastly, Group C includes horizontal indicators related to globalisation. These indicators include measures of economic and work life in the KBE. There are indicators of trade, knowledge production and diffusion, economic structure (change) and human resources (especially international flows of HR) emerging in and/or driving the KBE. 3 Another example, the use of ICTs in government can also be considered as measures of public sector innovation. The provision of government services on-line represents an innovation in terms of delivery of services and increased transparency of processes. 4 Empirical research shows that such demand is highest among the youth cohort (persons aged 15 to 29). 2 KEI-WP2-D2.5

7 Some indicators are non-controversial and cover core information, both on their own and as critical inputs for composite indicator development. Other indicators expand the concept of a KBE, due to a need to move beyond the traditional core indicators. This is particularly the case when considering human potential measures, for example. Care must be taken in the construction of composite indicators. If they are poorly constructed or misunderstood, they can be inappropriately used and result in mixed misleading messages to policy. The OECD cautions: In the context of policy analysis, indicators are useful in identifying trends and drawing attention to particular issues. They can also be helpful in setting priorities and in benchmarking or monitoring performance. The composite indicator should ideally measure multi-dimensional concepts that cannot be captured by a single indicator alone, e.g. competitiveness, industrialization, sustainability, single market integration, knowledge-based society, etc. (OECD, b). While indicators may not provide a complete picture (e.g. GDP per capita can be a poor measure of societal well being), composite indicators in turn can cover up problems with the underlying indicators. Data timeliness and trends are a particularly important factor for the indicators. Detailed insights into sectoral innovation performance are essential for the development of effective innovation policy at regional, national and European levels. Many countries have good data series over time for the two main sectors of services and manufacturing trends, but what is needed is sector level data at the level where firms compete. This would require, at the minimum, disaggregated data at the two-digit NACE level. For some sectors such as pharmaceuticals and aerospace, data are required at the three-digit NACE level. For many indicators such as R&D and patents, data at a disaggregated sector level lags behind national data by several years. Sources of sector data include the OECD, some national statistical agencies, universities and research institutes. At Georgia Tech s Technology Policy and Assessment Center in the US, researchers have monitored high technology-based industrial competitiveness at the national level since 1986 and recently carried out analyses of high technology indicators (HTI) for 28 countries for the years 1993 to Sources of HTI indicators include the European Innovation Scoreboard, the Entrepreneurship Scoreboard, the World Bank, the World Economic Forum and the UN Commission on Science and Technology for Development (Newman et al, 2004). 5 5 There are HTI for Europe (UK, France, Germany, Netherlands, Italy, Switzerland, Sweden, Spain and Ireland), US, Canada, Japan, Australia and China. Sources for developing HTI for developed and emerging economies in a KBE are identified and include the OECD, the Innovation Scoreboard, the World Bank and the UN Commission on Science and Technology for Development

8 The dataset produced for KEI does not include disaggregated sector level data, mostly because these data are not yet consistently available for all types of indicators and all target countries. The Systematics Sectoral Innovation Watch project of the European Commission is developing sector level indicators from existing databases. It is also looking at the use of one-off datasets to compensate for information gaps. 6 Another policy priority for the European Commission is to develop indicators at the gender level. This includes indicators on the proportion of students in higher science programmes that are women, and the share of women working in ICT occupations. There is also a need for indicators on the changing roles for women due to transitions brought about by the KBE, such as improved job flexibility through ICTs. Good quality information on gender is available for educational achievements and for the number of researchers in FTEs. There is also good representation of gender breakdowns in labour force indicators (e.g. age, occupation) and mobility indicators (sector, occupation and international). Several interviews were conducted in 2007 to assess the policy and user needs for KBE indicators and to help identify new and improved indicators for emerging policy challenges 7. The interviews targeted policy makers inside and outside the EU - those involved with policy coordination, implementation or evaluation - as well as policy advisers in statistical agencies and academia. The results indicated that current indicators on innovation are not adequate from the user perspective, partly because of issues to do with timeliness, detail or international comparability, but partly also because of the changing needs for indicators in a KBE. Aspects of innovation most in need of indicator development included: Innovation flows (indicators on innovation capabilities, adoption and diffusion of innovations), Economic impact of innovation (indicators on the impact of subsidies or tax exemptions), Innovation collaboration (indicators on linkages, networks and clustering), Human resources (better indicators on mobility, gender, life-long learning, financing of researchers, graduate job markets), Social impact of innovation, Security issues (how do tighter immigration policies affect international HR mobility?), 6 These sectors are: Biotechnology, Food/Drink, Machinery/Equipment, Textiles, Chemicals, ICT/Electrical/Optical, Space and Aeronautics, Automotive. The horizontal topics are: eco-innovation and gazelles (fast growing SMEs). 7 This section draws from Arundel, A., van Cruysen, A. and Hansen, W., Policies for a Knowledge- Based Economy, KEI Deliverable 1.3. Maastricht: UNU-MERIT. 4 KEI-WP2-D2.5

9 Role of consumers in the innovation process, Venture capital, Organisational innovation, Changing nature of research: convergence of different scientific fields (e.g. how is this affecting the outputs from research?). Economic sectors in need of better indicators (and better classification systems) were identified as: Services, including indicators on e-government and e-health, among others, Biotechnology and nanotechnology, Energy and environmental technology, ICT (more indicators on use and impact in firms and households). Finally, the interviewees hoped for better indicators to measure the effectiveness and impact of innovation policies

10 2. INDICATORS 8 This section describes the main classes of indicators for a KBE. Each section concludes with a brief description of key indicators. Annex 1 gives additional information on each indicator, including the source, description, unit, time range for data availability, and the number of countries for which the indicator is available. The indicators in the Annex can be identified from their alphanumeric code. Many of the key indicators are traditional indicators for which data are relatively easy to find. There are also a few more recently developed indicators. However, the list of key indicators is conservative, with a strong bias towards familiar and well-understood indicators. Several sections below identify indicators that are currently not included in the key indicators, but which should be considered in the future to round out coverage of a KBE. 2.1 Group A: Characteristics and drivers Recent development work is ongoing at the national (e.g. National Science Foundation U.S.), EU (e.g. Eurostat), and international levels (OECD and UN). Both national and international agencies are addressing the need for new aggregations of relevance to the KBE. For example, under ISIC rev.3.1 there is now an information and communication technology (ICT) sector (Gault, ). The 4-digit aggregation of the ICT sector makes it possible to examine ICT sector growth and also makes it possible to compare ICT sector growth with other high tech sectors in the KBE. There are several initiatives for indicator development in Europe. NEWKIND aims to develop new indicators for the KBE looking at changes in economic, industrial and firm performance. The NIND project in Scandinavia is developing new indicators from exiting data sources, such as the European Community Innovation Survey (CIS). There are also other international cooperative efforts that can be tapped for indicators. One example is the KBE task force for Asian-Pacific Cooperation (APEC) that provides links to a range of comparable indicators for Australia, Canada, China and Japan (as well as countries not targeted in the KEI study such as Malaysia, Chile, Mexico, New Zealand, Philippines, Thailand and Singapore). The indicators are grouped according to business environment and include KBE indicators of government transparency; ICT infrastructure that includes indicators of new technology take-up, domestic telecommunication capacity (e.g. number of computers), internet users and e-commerce; innovation indicators of number of researchers, GERD/GDP, BERD/GDP and knowledge networks; and, human resources including science graduates and the human development index (APEC, 2002). 8 The indicator references - A1a, A1b and so on - refer to the indicator categories in the Excel dataset associated with this WP, as well as Table 1 in Annex I, which contains more detail on these indicators. 6 KEI-WP2-D2.5

11 2.1.1 A1 - Production and diffusion of ICT There has been extensive development of new indicators for high tech sectors, in particular indicators on the growth and economic and social impacts of ICT use. Most countries have ICT indicators, although the scope and coverage varies. Finland, the United States and Canada continue to lead the way in measuring ICTs and developing indicators for ICT growth and social and economic impact assessments. In the US, the Department of Commerce collects a large number of indicators on ICTs. In Canada, a programme was established at Statistics Canada to develop indicators for an information society/information economy, and as a result, there is a wealth of information available. The project e-europe collects and provides benchmark indicators for ICTs. For enterprises, there are indicators of access and use of the internet, cost of access to e- commerce, e-readiness and e-government. For citizens acting in e-europe, there are indicators for citizen access and use of the internet, internet access costs, e-government, e-learning, e-health and ICT-security. This is the set of indicators that have and continue to receive a great deal of policy and measurement attention. These are also the indicators that, although grouped together here under production and diffusion of ICT, may also be useful for other indicator areas A1a Economic impact of ICT sector Indicators for the economic impact of ICTs remain a top priority for policy and decision makers. The two key indicators for direct economic impacts are ICT value-added as a share of total business sector value-added and ICT employment as a share of total business sector employment. Unfortunately, the latter is not consistently available across most EU countries. An indicator for ICT sales as a share of total business sector sales is of less importance. Investment in ICT is a key indicator because it covers expenditures on ICT across all sectors and provides national indicators for the diffusion of ICTs, which is important to productivity (e.g. diffusion, adoption). Finally, patent data provide an indicator of national capabilities to develop ICT innovations. The key indicators included in the dataset are: A1a1 Share of ICT sector value-added, A1a3 Investment in ICT as a percentage of GDP, A1a4 Number of patent applications to the EPO in ICT patent classes per million population, A1a5 - Number of patent applications to the USTPO in ICT patent classes per million population

12 A1b to A1d Internet use In addition to economic indicators of ICT use, indicators for the diffusion of the internet to businesses and private citizens capture the potential for a range of new business models and services. Indicators on internet use by individuals also support human resources, skills and social policy information needs. This means that some of these indicators can also be used to develop composite indicators of citizens in the KBE. We might consider developing an index of skills, trust in the government and internet usage by individuals. We need to consider internet use (in general) by firms, government and individuals. A1b Internet use by firms Not all enterprises have access to broadband enterprises operating outside of urban centers and in the new member states may not have broadband access, but internet penetration rates are increasing rapidly. This means that this indicator has a short lifetime and is consequently not included in the list of key indicators. A better option is to produce indicators on how businesses use the internet. Developing indicators on firm internet usage for orders and sales is difficult. ICT indicators may not represent all sectors (e.g. accounting firms, consulting firms, legal firms) and will biased towards internet-based sales companies such as software or book retailers. How representative the measure of receiving orders over the internet is for small firms such as butchers and bakeries is another matter, which explains the division of the indicator into all businesses and SMEs. A better indicator for the future could be the share of businesses that maintain a website or homepage as a share of businesses that have access to the internet. Businesses receiving orders over the internet by NACE could be a third indicator. The key indicators included in the dataset are: A1b1 Businesses receiving orders over the internet All businesses, A1b2 Businesses receiving orders over the internet SMEs. A1c Internet use by individuals Indicators on internet usage by individuals present even more challenges. We can consider segments of the population either by age or activities (e.g. students). We also need to consider access before usage. This remains an underlying weakness of data on usage broadband access may not be available, or the cost may be prohibitive for households, and citizens could instead access the internet at work, school, libraries or other public sources. A key indicator for individual usage is therefore broadband access as measured by penetration rates. This is a key indicator because only broadband access enables the internet to be used for a range of productivity enhancing innovations for both the public and private sectors, such as completing tax forms over the internet or e- banking. Here, it would be useful to have urban and rural results separately, but this breakdown is not commonly available. Once we have an indicator of penetration based 8 KEI-WP2-D2.5

13 on the share of households with broadband access, we can develop indicators on usage through the type of activities the individuals engage in, such as e-banking, e-trading or e- education. Eurostat has data on individuals using the internet for specific purposes including sending/receiving s, finding information about goods and services, reading/downloading online newspapers, playing/downloading games and music and internet-banking. The key indicators included here are: A1c1 Percent of individuals making any use of the internet, by age, A1c4 - Broadband penetration rate. A1d Government ICT Public sector innovation is a particular policy interest. How is the public sector doing in terms of workplace innovation itself? The available data are very limited. This is in part due to the definition of public service innovation, i.e. what should be included. There is also a lack of data at the EU level. For these reasons, we can borrow from ICT indicators to provide information on government services online and the types of online services accessed by firms and private citizens. These are important indicators of public sector innovative practices. 9 Eurostat collects data under Information and Society activities that can shed light on public sector innovation in delivering services to business and private citizens. Although the data are based on individual and company surveys, they can be used to develop indicators on the types and volume of public government services based on usage. One important indicator is the percentage of individuals using the internet for interacting with public authorities (the data are available by purpose and include obtaining information, obtaining forms, returning filled in forms, and full electronic case handling). A second important indicator is the percentage of enterprises using the internet for interacting with public authorities, again broken down by purpose. There is a time series covering 2002 to. The only key indicator included in the dataset under this category is: A1d2 Percent of individuals who use the internet to interact with public authorities A2 - Human resources, skills and creativity Indicators for human resources include traditional indicators for scientists and engineers and cultural and policy support for technology based development and entrepreneurship. For the more traditional indicators of GERD, FTE researchers, doctorate degrees 9 For instance, innovation in delivery of services appears under ICT enabling technologies that make government services more transparent and more accessible. The filing of personal income tax and business taxes on-line tend to be captured under ICT measures rather than the fact that the development and adoption of the technologies within the public sector is innovative

14 awarded, and student interest in science and technology, there are comparable indicators for India and China. There is also a choice of sources, including national statistical agencies. The National Science Foundation of the US provides reasonable coverage for Japan, South Korea, Taiwan, China and India. The OECD is expanding its coverage of emerging economies, particularly China and India. The knowledge intensity of an industry can be examined by using the education and skills of workers employed in the industry. These can be compared with other industries in the country and also with the same industries in other countries. Other indicators include the rate of skill level increases, shifts in skills by field of education, the presence of knowledge workers, and relative earnings across knowledge intensive sectors and across countries (Baldwin and Beckstead, 2003). Indicators for skill levels are available for most countries, although the field of education has more limited availability and limited international comparability. The knowledge intensity of an industry can also be measured using education and output indicators. For example, one can link the scientific base of an industry to the education field of science by developing a concordance between knowledge output (e.g. patents, bibliometrics) and field of education. Work to develop indicators for the education base of selected industries in Europe (a set of pilot countries) is currently underway for the European Commission. The use of ISCED 1997 makes the international comparison of university-qualified scientists and engineers possible although there are still problems. The OECD is working with Eurostat to address this. Nevertheless, the international classification can be used to compare the educational level of skilled workers across all countries. It is possible to collect indicators to show the pool of skilled workers being produced by national systems, and for some countries it is possible to develop indicators of the foreign contribution of scientific and engineering students. National agencies keep track of foreign student enrolment, and so indicators can be produced on the potential supplies of knowledge workers from other countries. For example, there are indicators for the US on the plans of foreign graduates earning doctorate degrees and we can compare the plans of graduates from countries across Europe, as well as those in India and China. Are students from some countries more likely to stay and work in the US after graduation? Do students from some countries tend to return home after acquiring useful work experience in the US? This type of information is not available for many countries, although this will be addressed in the coming years as surveys of PhD graduates provide more extensive coverage for European and Asian countries. 10 The flow of knowledge workers in the KBE means the flow of ideas. For Europe, there is no one source of information about the process of mobility and flows between occupations, sectors or countries. One of the key sources is the European Labour Force 10 A number of countries carry out some form of graduate or destination survey. These surveys of graduates provide important indicators of mobility from education to the labour force, as well as indicators of flexibility of skills sets (e.g. occupation choice, sector choice) or demand for skill sets. 10 KEI-WP2-D2.5

15 Survey (LFS), which can provide some basis for comparison with the US and Canada. It is probably the only database that provides international comparisons over time. One of the weakest areas of indicators both in Europe and North America is that of lifelong learning. This increasingly catch all phrase has problems of data availability and comparability. The EU has carried out several surveys of employer training without much success. The best source is the LFS, which asks individuals about participation in training or courses. However, the definition of life-long learning remains unclear. The most common indicators for human resources typically focus on highly-skilled workers while paying less attention to lower levels of education. For instance, there are substantial efforts for indicators on PhD-qualified skilled workers and researchers but less attention on technically skilled personnel and undergraduates (e.g. bachelor degree). This narrow and dated focus on high skill level indicators make it difficult to develop indicators for a KBE in which creativity and job growth can be evaluated across occupations and skill levels, and most certainly below the PhD and higher skill levels. We need only look as far as ICT occupations, where a variety of skill levels and fields of specializations are present. This means that we need to develop indicators on a broader range of skill levels as well as information on specialisations. All in all, there are many questions surrounding human resources and skills including the supply (e.g. education pipeline), factors that may influence the pool (e.g. mobility) and creativity (e.g. life-long learning and the ability to move in a job market that no longer provides good job security). This means that we need indicators for the trends for postsecondary education, but we also need to know what happens to the graduates. While we need to know the representation of researchers in the work force, we also need to know of the exit of researchers brought about by career path changes and international mobility. Indicators need to be developed to consider: General education and trends such as literacy and numeracy, Specific skill trends such as the number and types of researchers, Factors that might influence supply, including various forms of mobility from occupation-to-occupation, sector-to-sector and country-to-country, The potential for work place innovation and change as measured by training and lifelong learning, The creativity of people as measured by tolerance and acceptance fostered through social trust and the share of creative occupations A2a - General education indicators With regards to the overall standing of a population and its ability to benefit socially and economically in the KBE, we need to know about the basic skills for learning and development potential. For national and international comparability, literacy scores are

16 key indicators of a population s educational health. While there has been a tendency to focus on mathematics and/or science literacy, it would be a mistake to leave out reading literacy. The changing nature of jobs (e.g. the growing need for communication and networking skills, writing skills, and marketing skills) tells us that indicators for reading comprehension, as gathered for 15 year olds in PISA 2000, should be considered as important as mathematical literacy (PISA 2003). Reading literacy is excluded from the key indicators because the data are increasingly dated, but this indicators should be included when next available. It would be useful to have high quality, comparable indicators for reading and mathematical literacy among adults, but none are available. PISA 2006 will provide measures of science literacy that should be included in future indices. Indicators on graduation rates by level of education and field of qualification and gender are important. We need to be able to develop composite indicators of education pipeline output to the population in general, the labour force and for considering issues such as representation of men and women in fields of S&E and S&E occupations. Eurostat has this data available on a timely basis for the EU countries and the data are also available for the US, Japan, and Canada. The minimum requirements are indicators for level of education ISCED 5A and 5B (all tertiary education from a one-year diploma to completion of a bachelor s level degree) and ISCED 6 (PhDs). This represents a change to the indicator under A2a4 (see below), as we need to have breakdowns for undergraduates as well as PhDs. We can then consider the share of the population (whether employed or unemployed) with a university qualification. 11 General education by field of study can be included under HRST (human resources in science and technology) indicators under A2b5. The most useful indicator by field of study is to express specific fields as a share of the total. Too often there is a focus on output in terms of the number of graduates, even though the specialization is slipping in terms of the share of total degree output. The key indicators included here are: A2a1 Pisa mathematical literacy of 15 year olds, A2a4 New PhDs per thousand population aged 25-34, A2a5 Percent of working age population (15-64) with a tertiary education (ISCED 5 and above), A2a7 Percent of adult population (25-64) with an earned doctorate degree (ISCED 6). 11 With graduates by level as the numerator, and the overall population as the denominator, we have a measure of the education of the adult population; with the number as graduates as the numerator and the labour force as the denominator we have a measure of the skill levels of the labour force. Both are important education indicators, one presenting a broader picture for public policy and one more refined for labour force and immigration policy. If resources are short then we should go with labour force as the denominator because the PISA results provide indicators of general education for the general population. 12 KEI-WP2-D2.5

17 A2b - HRST education indicators It is important to have indicators on HRST in the work force and this will include a variety of skill levels. The number of researchers per 1,000 labour force is a traditional indicator and remains in demand by policy. It is linked to other R&D investment measures. The only indicator included in the dataset under this category is: A2b1 Total HRST as a percentage of the working age population A2c - Skills Developing indicators on skills has always been a difficult task especially when one moves beyond ISCED. Formal postsecondary qualifications and indicators of the upper skill levels are still the best choice, due to data availability. Measures and indicators on other skills of relevance to a KBE are also available for some countries, such as nontraditional certification of specialists and graduates of e-training programmes. Part of the change brought about in the KBE and enabled by ICT is the onus on employees to be responsible for their own training programmes, as companies retreat from their responsibility to train and/or provide training for their employees. For some enterprises a new piece of software is the extent of training. Under the theme of life-long learning, Eurostat maintains a database on enterprise training activities. We should consider an indicator on training enterprises as a percent of all enterprises and we could develop composite indicators by considering the type of training. If resources allow, we should include an indicator on training by NACE. Life-long learning and hours spent on training are important indicators. At this time, the key indicators included in the dataset are, however, limited to the following: A2c2 Average hours employed persons spend in training, A2c3 Participation in life-long learning per working age population (25-64) A2d - Creativity Highly trained and educated individuals are not the only focus of human resources in the KBE. The ability to create and combine knowledge involves much more than formal education and qualifications. We therefore need indicators on the creative capacity or potential of people. In order for persons to think and work in a creative manner, particularly to take risks and pursue new thinking that may lead to innovation, a person must have a certain level of comfort or trust in his/her environment. For this indicator the European Values Study Survey, a large-scale longitudinal survey that includes measures

18 of social and political values, is useful. The data can be used to develop indicators of social and political trust. As mentioned earlier, it is important to develop indicators that go beyond the traditional indicators. This includes considering the various human resources in the innovation chain that contribute to creativity and not just researchers. It also means considering occupations beyond the core S&T occupations and the need to consider creative occupations including ISCO 1 (senior managers and officials), ISCO 2 (professionals such as scientists and researchers) and ISCO 3 (technicians and associated professionals), all part of the chain of processes of innovation. The WP2 dataset of indicators does not include any creativity indicators A2e - Mobility The mobility of scientists and engineers is commonly believed to diffuse knowledge and consequently support innovative capabilities, but indicator development remains weak. There are two dimensions to the policy interest in mobility. There are policies to promote internal mobility where the situation is seen as win-win, benefiting all parties. Examples include programmes to encourage mobility within Europe and hence knowledge exchange, or programmes to facilitate public sector and private sector exchanges to encourage knowledge diffusion. Government-university interchanges encourage networking and knowledge exchange. European policy has identified the need to improve the commercialization of R&D and so policies that encourage and enable commercialization are important. Then there is mobility that benefits one party but may cost the other party. Flows from public research to private research due to lack of R&D support may undermine public research needs. A key policy interest concerns skilled researchers moving outside of the EU to carry out their research and the apparent lack of attraction the EU has for highly skilled foreign researchers. Although the data may be less reliable than desired, it is still important to develop some indicators on international flows. The issue of international mobility and immigration patters is taken up in more detail under C1 (globalisation). Many of the available mobility indicators are based on anecdotal evidence and ad hoc surveys, although these indicators are critical for R&D policy needs from education to immigration to R&D investment. The only indicator included in the dataset under this category is: A2e1 Job to job mobility by NACE. 14 KEI-WP2-D2.5

19 2.1.3 A3 - Knowledge production and diffusion Surpassed only by the madness for indicators and measures for ICTs over the last decade, indicators for R&D continue to flourish and expand in country coverage and international comparability. The key sources for R&D indicators are national statistical agencies, Eurostat and the OECD. There are R&D indicators of some shape or form for every country, although the time series vary. Coverage of emerging economies is improving A3a - R&D family of indicators Indicators grouped under A3a represent the traditional R&D family of indicators. There are time series from 1981 to and good country coverage. The key indicators are as follows: A3a1 - GERD expenditure/gdp, A3a3 - GERD per capita population, A3a4 - Estimated civil GERD as a percentage of GDP, A3a5 - GOVERD as a percentage of GDP, A3a6 - HERD as a percentage of GDP, A3a7 - GBAORD as a percentage of GDP, A3a8 - BERD as a percentage of GDP, A3a10 - BERD as a percentage of value-added in industry, A3a11 - BERD/GERD A3b - Patents Patents measure invention. Some of the data on a patent application or grant can be used to measure the diffusion of knowledge. The KBE brings new challenges to the value and application of this traditional group of indicators. There is increasing collaborative patenting across national boundaries. Patent indicators are used by policy analysts as a measure of technology, knowledge and R&D outputs. But patent indicators are of limited value to innovation because not all innovations or inventions are patented and patent propensity rates vary substantially across sectors. Sometimes patents are not sought because the inventor wishes to maintain 12 The indicator on business expenditures on R&D as a share of total R&D expenditures (A3a11) is an indicator identified as important for the Lisbon agenda. This indicator is, however, limited because it does not report on efforts of European R&D firms in terms of performing R&D but only on European firms expenditures on R&D

20 secrecy. Patenting activity is even more concentrated than R&D. In 2002, only five countries (France, Germany, Japan, the United Kingdom, and the United States) accounted for 84% of triadic patent families compared with 78% of R&D expenditures (OECD, a). For the KBE, it is important to consider the innovative process at the international level and to use triadic patents, in which the same patent is granted by the USPTO, the EPO and the JPO. The advantage of triadic patents is that they capture the most economically valuable inventions and overcome the home country advantage (for instance European firms are more likely to patent in Europe than American firms, and American firms are more likely to patent in the United States). The disadvantage of triadic patents is that they lag other patent data by two years. The key indicators included in here are: A3b1 EPO patent applications by priority year per million population, A3b2 EPO high tech patent applications by priority year per million population, A3b3 USTPO patent grants per million population, A3b4 USTPO high tech patent grants per million population, A3b5 Triadic patent families by priority year per million population A3c - Bibliometrics The third group of indicators for knowledge production and diffusion are bibliometrics, or indicators for scientific publishing and citations. They can be used to develop indicators of scientific effort and knowledge of a nation. The National Science Foundation (NSF) produces a time series by country for several key indicators including: S&E articles by country, per capita output of S&E articles by country, and country portfolios of S&E articles by field of specialization. We may wish to examine an indicator with S&E article output by specialization as the numerator and the number of S&E degrees by specialization as the denominator. The data are available online at The key indicators do not include any bibliometric indicators A3d - Knowledge flows The fourth family of indicators looks at knowledge flows in terms of firms and public research institutes and universities. The value firms place on knowledge flows with public research institutes is an important measure of the value of public research to innovation and the ease at which firms can access and commercialize this knowledge. Knowledge flows among firms is also an important indicator which can be developed by examining the share of firms collaborating on innovation. Here firm size would be useful. The key indicators included in the dataset are: 16 KEI-WP2-D2.5

21 A3d1 Share of all firms reporting public research with universities as a high value information source, A3d2 Share of all firms reporting public research with institutes as a high value information source, A3d3 Share of all firms reporting public universities as a research cooperation partner, A3d4 Share of all firms reporting public institutes as a research cooperation partner, A3d5 Share of SMEs collaborating on innovation A3e - Total investment in intangibles The last group of indicators under the drivers of a KBE considers investment in knowledge as a share of GDP. For this indicator, data on R&D, higher education and investment in software are combined. Another indicator is the value added of knowledge intensive exports expressed as a share of total exports. The NSF, in its 2006 Indicators Report, provides a table of data for 1980 to 2003 showing world industry value-added by industry. There is also a table on exports/imports by industry that would augment the Eurostat data. The only indicator included in the dataset under this category is: A3e1 High tech exports/total exports A4 - Innovation, entrepreneurship and creative destruction Innovation is widely believed to be central to European economic competitiveness and is consequently featured in a range of policy documents. Many pro-innovation policies are based on subsidizing knowledge production, either directly or through fiscal benefits for R&D, but innovation is also supported through manufacturing extension services to promote the adoption of productivity enhancing technologies such as Advanced Manufacturing Technology (AMT) or ICT, other information services, and subsidies to SMEs for hiring scientists and engineers or acquiring new technology. A few countries also provide government support for innovative start-ups, such as venture capital. Entrepreneurship is difficult to support, but there is widespread discussion about including courses on entrepreneurship and patenting in the syllabus for science and engineering students. Entrepreneurship is also supported directly through government provision of grants, loans and early-stage financing to small innovative firms and indirectly through tax relief to investors in venture capital firms

22 A4a - Entrepreneurship Our interest in entrepreneurship is due to its role in the creative destruction brought about by innovation in the KBE and the ability of countries to respond efficiently, both by moving into new areas of economic opportunity and to replace inefficient producers. The churn for enterprises from the upheaval in activities driven by ICTs and globalization can be measured by the entries and exits of firms. The entries and exits reflect births and deaths of firms, as well as relocation. Rather then consider firm entries and firm exits apart, it makes most sense to develop an indicator of churn by using both data sets. The references in the dataset for these two indicators are: A4a2 Firm entries, A4a3 Firm exits A4b - Demand for innovative products The share of the population aged is a good indicator of potential demand for innovative products. This is the age group that quickly adopts new products and consumer demand drives innovation in products. At the other end of the scale is the need to have indicators on the potential demand for new products and technologies by the older cohort, that of persons aged 65 and over. One indicator of comfort with new technology and potential for increased demand is the use of the internet and ICTs. As part of its survey of the internet for specific purposes, Eurostat provides data by age cohort. For projecting demand for innovative products, we could also construct an indicator with below 65 age cohorts together with population projections and usage to estimate demand for internet and related technologies. We also need to consider indicators for other activities that could drive the need for innovative products. For example, the ageing of the population could accelerate the use of the internet by the older age cohorts for e-health, just as life-long learning can generate increased demand for innovative products through e-learning and e-communities. Concerns about the environment and energy could also drive e-commerce and e-working (e.g. teleworking; conference calls in lieu of face-to-face meetings). Another indicator for the demand for innovative products is to develop an index on the sophistication of local buyers. This is the best indicator for demand for innovative products and is available from WEF,. The purchasing power as measured by per capita GDP is another important indicator. The government is an important driver of innovative products given its appetite and budget for work reorganization and change. Government procurement of advanced technology products is therefore an important indicator of demand. Firm level absorption is another key indicator for potential demand for innovative products and processes. 18 KEI-WP2-D2.5

23 We suggest the current key indicators include: A4b3 GFCF (cross fixed capital formation) as a percentage of GDP, A4b7 Firm level technology absorption A4c - Financing of innovation Indicators on venture capital remain one of the weakest groups of indicators for measuring innovation. The NSF publishes a time series (1980 to 2004) on US venture capital disbursements by industry, by financing stage from early stage to seed stage to later stage and by venture capital seed disbursements. Unfortunately, no such data exist at the world level. However, there are some data available via and while these data are not the most reliable, they can provide us with a rudimentary indicator on venture capital financing in the early stage. We have not included indicators in our dataset for this category A4d - Innovation sales share We have a set of indicators to look at share of total sales from innovative products. Here we consider the share of firms introducing new products to the market and sales from the new to market products. The data for these indicators come mostly from the CIS: A4d1 Share of firms introducing new-to-market products, A4d2 Share of total sales from new-to-market products, A4d3 Share of firms introducing new-to-firm products, A4d4 Share of total sales from new-to-market products, A4d5 Community design registrations per million population A4e Organisational innovation 13 Organisational innovation can be a key factor in both productivity growth (Brynjolfsson and Hitt, 2000; Murphy, 2002) and the ability of firms to profit from product and process innovation. The fourth CIS provides new options for constructing indicators on organisational innovation on 1) knowledge management, 2) organisation of work, and 3) relationships with other firms or public institutions. CIS-4 also asks about four types of outputs of organisational innovation, so it is possible to determine which of the three types of organisational innovation have the greatest impacts on quality or production costs. A main limitation for CIS indicators on organisational innovation is the lack of time series. 13 Organisational innovation is covered by WP4 on missing indicators and future indicator needs